Hive ,row_number,dense_rank,percent_rank不要带参数；rank要带参数

最新推荐文章于 2023-02-26 23:22:25 发布

二十六画生的博客

最新推荐文章于 2023-02-26 23:22:25 发布

阅读量1.9k

点赞数 4

本文链接：https://blog.youkuaiyun.com/u010002184/article/details/109499535

版权

Hive Spark Hadoop MapReduce 专栏收录该内容

162 篇文章 ¥29.90 ¥99.00

订阅专栏

超级会员免费看

本文介绍了在Hive 2.0.0版本中，row_number(), rank(), dense_rank(), percent_rank()等窗口函数的正确用法。在使用时，row_number(), dense_rank(), percent_rank()不应带参数，而rank()函数需要带参数。示例展示了如何在partition by后根据request_time进行排序。" 120036497,11386727,深入解析Android LiveData：原理与生命周期交互,"['移动开发', 'Android开发', 'LiveData', '生命周期', 'MVVM']

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

正确如下：

row_number() OVER( partition by session_id order by request_time) as row_num -- 不能带参数
rank(request_time) OVER( partition by session_id order by request_time) as row_num -- 要带参数
dense_rank() OVER( partition by session_id order by request_time) as row_num -- 不能带参数
percent_rank() OVER( partition by session_id order by request_time) as row_num -- 不能带参数

错误如下：

row_number(request_time) OVER( partition by session_id order by request_time) as row_num
rank() OVER( partition by session_id order by request_time) as row_num
dense_rank(request_time) OVER( partition by session_id order by request_time) as row_num
percent_rank(request_time) OVER(

了解本专栏